The observation that risk of Type 1 diabetes is linked to major histocompatibility complex (MHC) and human leukocyte antigen (HLA) genetic loci was an initial clue that this condition was an autoimmune disorder. Further work has identified immunodominant peptides including an insulin B:9-23 peptide a GAD:524-543 peptide that binds to the high-risk class II MHC I-A(g7) in the NOD mouse. What has been lacking, however, has been information about the T cell receptor (TCRs) of these peptide specific clones. The proposed experiments of this grant were designed to test several related hypotheses that relate to several important and unanswered questions of autoimmunity, including the role of susceptible and protective MHC, the observation of autoantigen spreading, the stochastic occurrence of disease in genetically identical animals, and the activation of only certain subsets of autoreative cells in disease prone mice. Experiments were designed to test several related hypotheses: (1) an autoantigenic peptide/high-risk MHC complex is recognized by T cells with the most "permissive" TCR pattern; (2) this same autoantigenic peptide, presented by other, non-disease related MHC molecules, is recognized with a different and more "complex" TCR pattern; (3) a dominant chain from autoreactive T cells with a "permissive" pattern of recognition can pair with an irrelevant opposing chain and still maintain reactivity to the autoantigenic peptide; (4) a single (alpha, beta) TCR can potentially recognize two very different peptides presented by the same MHC molecule; and (5) finally, insulin B:9-23 reactive T cells containing the V(alpha)13.3a segment are critical for the initiation or progression of autoimmune diabetes in the NOD mouse. These questions will be tested by immunizing mice that have different class II MHC (NOD/I-A(g7), NOD-I-A(h), C57BL/6) with a series of defined islet peptides (insulinB:9-23, insulin A:7-21, and GAD:524-543) and control peptides (tetanus toxin and KLH). The T cell receptor chains will be sequenced and the TCR patterns from the different panels will be compared. DNA constructs will be created and inserted into TCR shuttle vectors and inserted into a T cell hybridoma that lacks an endogenous TCR to create a series of T cells with hybrid T cell receptors. These cells will be tested for reactivity to peptide. Several experiments will attempt to isolate or create cells with hybrid receptors that have dual reactivity to two different peptides. Preliminary data suggests that a single TCR with an insulin reactive alpha chain and a GAD reactive beta chain could recognize both autoantigenic peptides. Finally, young NOD mice will be immunized with either V(alpha)13.3a TCR peptides selected from the polymorphic CDR regions or V(alpha)13.3 DNA constructs, in an attempt to eliminate or immunomodulate endogenous V(alpha)13.3a containing T cells that are potentially autoreactive.